Sage Advice Compendium Update 1/30/2019

[DND_Reborn]

First, I said it was a "nearly pointless feature" [of the feat], not that the feat itself was pointless without this. That is an important distinction. (bold added for emphasis)

From a power perspective there are a lot of pointless feats. I don't feel feats were well balanced in terms of power. They feel more about flavor than power to me. As such you get some utterly weak ones and some irreplaceably powerful ones.

That said knocking someone prone is still potentially beneficial to your allies and slows down the enemy from getting to your squishies. So even in the worst case interpretation, shield master isn't a pointless feat.

By the way consider this strategy in a 1v1.

Fighter 1: Attack, attack, shield bash Knock prone. move 20 ft away, taking an OA at disadvantage.
Fighter 2: Stand up, dash 20 ft ends turn adjacent
Fighter 1: repeat.

There's plenty of benefit to shield master in 1v1 combat even when you can only shove after. You just have to think tactically.

As I said, unless you have allies who can take advantage of the knock down you caused, the opponent can stand up on their turn before you benefit from the knock down. Now, if you want to keep allowing OA at disadvantage against you, I suppose that would be a way to deny them any real attack. There is nothing wrong with that and I hadn't thought of it, so it is more useful when used defensively. However, it offers you no benefit offensively otherwise and if you look at the rest of my post, it demonstrates a perfectly acceptable way to use it where it at least can be used offensively by you without the need for having an ally there to hit the target.

So, thanks for the idea on how to use it defensively at least, that does give it some more merit even as currently ruled.

5e system is built and designed around team play, do the assumption is there will be allies and elements are based on it.

Have you looked at how effective bardic inspire dice are sat levrls 1-2 if you are solo? How effective Inspiring leader is solo? Or the help action entirely solo?

Shield Master gives you three benefits, two of which are solo fine - they just affect you. It cannot have one with diminished capability while solo?

Now for your example, let's take two fighters at 5th level one-on-one.

Stan has Shield Master, Dave does not. Both are primarily sword xnd board types.

Stan takes two swings, 5th level then bonus shoves.
Say Dave goes down.
Stan steps away allowing Dave a disadvantaged AO. Moves 30 ft.
Dave can get up, but cannot close to melee this turn. Maybe he can juggle the interactions to throw something but that's hardly a match.

So one disadvantaged and one interaction juggled throw vs two normal melee swings.

"nearly a pointless feature"?

Now swap Dave for any number of beasts/monsters with less than 50' movement and reach combined - now they likley dont have a "throw option." Now every round you can bonus shove and succeed at the end is an exchange where you gain the edge of taking your two melee attacks at full and all the opponent gets is one disadvantaged AO.

"nearly a pointless feature"?

As one-third of a feat?

Even after limiting it to solo fights and ignoring its team play potential?

Gotcha. Sure thing.

Except Dave doesn't move to engage. He gets up and readies his attack for when Stan moves into his reach. Now, when Stan moves to engage Dave again, Dave gets his attacks before Stan. Also, you can at best assume a 50-50 chance for the knock down. As well, you're ignoring any feat Dave might have that would assist him, perhaps Sentinel, so if he does his with the OA, Stan isn't moving any more. At any rate, as I expressed above, this is a useful "maneuvering" way to employ Shield Master, which I thanked FrogReaver for.

As for your TWF run on...

" It is illogical to assume every time the character attacks they MUST strike with the main-hand first."

Yes, exactly, which is why "main hand" and off-hand are not in the 5e TWF rules. There is no limitation that requires either hand to start the sequence, just that the bonus action must be with a different light weapon in a different hand. That's all. Depending on styles and specifics, that bonus attsck may have no ability modifier but that's tied to its bonus action snd could apply to one hand on one turn and the other on the next.

Far as I can tell, your shortsword dagger whatever examples are all legal within std 5e rules and the only change your tule fir was allow a little different answer to "which attack loses the ability score bonus" **if** they lack the style that negates that.

Not really mind blowing?

Cool, then no worries, but you could be less of an ... about it.

 

[Asgorath]

Maybe, but I don't find that a very compelling case. Unless there's something very important I'm missing, it would seem most likely IMO that all actions are either instantaneous or all actions last for their duration. I've proven they can't all last for their duration. Therefore my conclusion is that they are all instantaneous. You can say what if, might and maybe all you want but at the end of the day I need more evidence than mights and maybe's to change my mind.



So you posit Natural Explorer as a counter example for why actions can't all be instantaneous. Let me take a look. Oh I see. It's not a capital Action like the attack Action or the dodge Action etc. That makes my answer there pretty simple. Since it's not an Action then it's not instantaneous.

You bring TWF in as well. TWF also requires a trigger of an attack. An attack is not an instantaneous event like an Action is. So it can trigger off an attack just fine and not contradict my stance.

You bring up the notion that shield master bonus action shove must be taken after the attack action is completed. I agree. However, I don't believe the attacks are part of the Action itself but rather that the Attack Action is instantaneous and taking it grants you the ability to make attacks.

So I think I'm covered against all those counterexamples. Got any thing else?

Let's be very clear: I am not arguing that actions last for their duration, I've been arguing the exact opposite this whole time. I'm just really confused how you get from actions resolve instantly to "oh you can Shield Master shove any time you like on your turn".

The Attack action happens the instant you make an attack (i.e. there is no declaring you'll take one later on your turn). This trigger condition is now true, and thus grants bonus actions from features like Shield Master, TWF, Martial Arts and so on. I'm using the Natural Explorer language to drive home the point I've been making about the "If you X, you can Y" sentences that are throughout the rules, where these very clearly mean that X has to happen before Y can happen (i.e. Natural Explorer makes no sense if you just say "Oh yeah I'll do Y (move stealthily at full pace) now, because I'll do X (travel by myself) later").

So, it seems like the sticking point here is that you believe the Attack action is a declaration, at which point the DM hands you one or more tokens that say "you can make a weapon attack", and you can trade those in to actually make an attack. The DM then takes the tokens back at the end of your turn. I just don't believe the Attack action works that way. The PHB says:

Attack

The most common action to take in combat is the Attack action, whether you are swinging a sword, firing an arrow from a bow, or brawling with your fists.

With this action, you make one melee or ranged attack. See the "Making an Attack" section for the rules that govern attacks.

Certain features, such as the Extra Attack feature of the fighter, allow you to make more than one attack with this action.

Highlighted the part that I think is important. Can you show me where in this text it says you can declare you're taking the action and make the attacks later? I'm just not seeing any evidence to support this, and my reading is confirmed by JEC:

https://twitter.com/JeremyECrawford/status/995319563523784704

"D&D combat is sequential, with no action-declaration phase at the beginning. Your turn can also be interrupted by someone’s reaction. Such an interruption could, among other things, incapacitate you, meaning your intention to take a certain action was never fulfilled."

The Attack action is different from Dash, Dodge, Disengage and so on. Dash simply doubles the amount of movement you have on your turn. Disengage can absolutely be read as the DM giving you a token that says "your movement no longer provokes OAs" which they take back at the end of your turn. Dodge can absolutely be read as the DM giving you a token that says "attacks against you have disadvantage" that they take back at the start of your next turn. These all have very clear timing in the text, specifically talking about when the effect ends. The Attack action has no such language, so I'm not following how you think this lets you just take your attacks at some point in the future.

As always, Extra Attack complicates this, but there is a specific rule to deal with Attack actions that allow multiple attacks:

Moving Between Attacks

If you take an action that includes more than one weapon attack, you can break up your movement even further by moving between those attacks. For example, a fighter who can make two attacks with the Extra Attack feature and who has a speed of 25 feet could move 10 feet, make an attack, move 15 feet, and then attack again.

This very clearly allows you to insert movement between attacks of the Attack action. Some bonus actions have the trigger of "the Attack action and making a weapon attack" which means you have the bonus action after the very first attack, and don't have to wait until all attacks are taken. Some bonus actions have the trigger of "the Attack action", which the movement rule strongly implies is all the attacks combined.

 

[Asgorath]

Well, if a bonus action has an actual timing requirement I think it would almost certainly be based on an objective activity, like making a melee weapon attack, rather than simply upon a formal Action, like taking the Attack Action on your turn. Even if you interpret Actions and Bonus Actions to be handled concurrently, the objective activities they grant will be sequential. Take the bonus action attack from two-weapon fighting as an example: it uses almost the exact same language as shield master, but adds "and make a melee weapon attack" to it. The Attack Action happens "on your turn," but the melee weapon attack happens at a distinct point in time within your turn, as part of a sequence of events.

Please show me the text in the PHB that supports your first sentence here.

D&D is a turn-based game. Your turn is sequential, and is made up of discrete elements. The two main elements are movement and your action. The PHB defines 10 standard actions you can take. It provides rules that say you can break your movement up so that some happens before your action, and some happens after your action. On your turn, you take these basic building blocks and put them together in order, and that defines what you do on your turn. There are special rules for when your action consists of more than one event, such as moving between attacks from Extra Attack. The rules are full of triggers, and these triggers rely on a condition. Once that condition has been fulfilled, the triggered event can happen.

I see no evidence of hand-waving away what an action means, actions are the fundamental elements or building blocks of your turn. The Attack action, by default, is making a single attack with your weapon. So, by default, you might have 3 blocks: move, attack, move. That ordering of blocks defines what you do on your turn. You have an ability that grants you a bonus action when you take the attack action, allowing your turn to have 5 blocks: move, attack, move, bonus action, move (in that order). Extra attack also allows 5 discrete blocks: move, 1st attack, move, 2nd attack, move. The middle move is explicitly allowed in the rules text. TWF's bonus action block has a trigger of the Attack action and making a single weapon attack, so you can do: move, 1st attack, move, TWF attack, move, 2nd attack, move. The TWF block can come any time after the 1st attack block. Shield Master's bonus action block has a trigger of the Attack action, and thus must come after the attack(s). The only thing that can be inserted between the attacks is movement, per the rules, giving us: move, 1st attack, move, 2nd attack, move, Shield Master shove, move. There is no other legal ordering of these basic building blocks.

This is all very simple and very logical. There are no nested or concurrent actions, because the blocks must simply be arranged in order. A triggered block must come after the block that triggers it. No hand-waving away what actions mean, the actions are explicitly defined in the rules as the things you can do on your turn.

 

[Hriston]

Of course event's for the character happen in sequence. That's not what I'm asking.

In my games players sequentially declare and describe what they will have their character do. The character performs those actions immediately as the player describes them and in the sequence he describes them. That's why in my games I check right when a character is going to perform some action that he is actually capable of doing that. So in my game a player declares bonus action shove and I check to see if he's met the condition it requires of taking the attack action on his turn and he hasn't yet and so his declaration is then disqualified.

If in your games the player can declare a bunch of things all at once like I'll bonus action shove and then attack that guy (i'd describe this as concurrent) and then you check to make sure all the conditions on those actions were met and since you are treating the declaration as current then it would be. Then you have the character sequentially perform the described actions then I can understand the different of opinion.

If that's what you are doing then that at least makes sense to me, although it still seems like a bit convoluted process to me but the logic behind the interactions at least make sense even if I don't understand the logic for why you would play that way in the first place.

Let’s see if I understand your question this time. In your games, within each player’s turn, there’s a direct correlation between the chronological order in which action-declarations are made and resolved at the table, and the sequence corresponding events are considered to have occurred in the fiction, and you’re asking if the same kind of thing happens in my games. Do I have that right?

I’d say it mostly does. A player might plan out his/her entire turn ahead of time, but then each fictional event is resolved in the order it's considered to be happening in the fiction, so I don’t think there’s too significant a difference between our games in that respect.

Where I think the difference lies is in how meeting a condition which must be met "on your turn" is handled. You seem to consider checking for character capability in respect to meeting such conditions to be part of resolving the player's action-declaration, and that if such conditions have not been met at the table at the time of declaration/resolution, then the action-declaration is ruled to be impermissible. Whereas, in my games, the fictional action of shoving a creature is considered on its own with respect to resolution and whether it's permissible as an action-declaration, and the question of whether a condition is met "on your turn" is left open until it can be determined whether, in fact, the condition was met "on your turn", which, at the latest, is at the end of your turn, after all actions for that turn have been declared. In that way, I suppose you could say there's a type of concurrency between the declared actions of a player's turn when regarded in this way, and that's because the condition applies over the period of time of the player's entire turn, so all actions are considered together. This is in contrast to the opposing view that the condition must be met at the time of action-declaration/resolution. Does that make sense to you?

 

[Asgorath]

Let’s see if I understand your question this time. In your games, within each player’s turn, there’s a direct correlation between the chronological order in which action-declarations are made and resolved at the table, and the sequence corresponding events are considered to have occurred in the fiction, and you’re asking if the same kind of thing happens in my games. Do I have that right?

I’d say it mostly does. A player might plan out his/her entire turn ahead of time, but then each fictional event is resolved in the order it's considered to be happening in the fiction, so I don’t think there’s too significant a difference between our games in that respect.

Where I think the difference lies is in how meeting a condition which must be met "on your turn" is handled. You seem to consider checking for character capability in respect to meeting such conditions to be part of resolving the player's action-declaration, and that if such conditions have not been met at the table at the time of declaration/resolution, then the action-declaration is ruled to be impermissible. Whereas, in my games, the fictional action of shoving a creature is considered on its own with respect to resolution and whether it's permissible as an action-declaration, and the question of whether a condition is met "on your turn" is left open until it can be determined whether, in fact, the condition was met "on your turn", which, at the latest, is at the end of your turn, after all actions for that turn have been declared. In that way, I suppose you could say there's a type of concurrency between the declared actions of a player's turn when regarded in this way, and that's because the condition applies over the period of time of the player's entire turn, so all actions are considered together. This is in contrast to the opposing view that the condition must be met at the time of action-declaration/resolution. Does that make sense to you?

Aside: there's an official ruling the Sage Advice Compendium that this isn't how the game is designed to work.

Can you apply this timey-wimey logic to the Ranger's Natural Explorer, specifically this sentence:

"If you are traveling alone, you can move stealthily at a normal pace."

This sentence has the same structure as Shield Master's bonus action, specifically "if you X, you can Y". You're stating that you can Y, as long as X is eventually true, right? So, why can't I use this rule to stealth at normal pace whenever I want, because I declare that I'll travel alone at some point in the future?

Or, let's use another example from Shield Master itself:

"If you are subjected to an effect that allows you to make a Dexterity saving throw to take only half damage, you can use your reaction to take no damage if you succeed on the saving throw, interposing your shield between yourself and the source of the effect."

Let's say I get hit with a Cone of Cold, and make the CON saving throw. Why can't I use the "Y" portion of this (use my reaction to halve the damage) because I declare that I'm going to get targeted by an effect that allows me to make a DEX save for half damage later on my turn (i.e. the "X" portion)? If there's no strict timing requirement between X and Y like you're claiming, then I should be able to just put my shield in front of any effect that has me make a saving throw and take no damage when I succeed. At the end of the turn, how do we resolve this Schrodinger's Reaction? I can't just go back and turn a bonus action into an action here, like you're suggesting we do with the Shield Master shove case.

 

[Hriston]

What rule requires you to say whether you shove a creature as an action or a bonus action?

This one:

"You can take a bonus action only when a special ability, spell, or other feature of the game states that you can do something as a bonus action. You otherwise don't have a bonus action to take."

I don't think that answers my question. I didn't ask what rule tells you whether you can shove a creature as an action or a bonus action.

Besides, the Shield Master feat states I can shove a creature as a bonus action. If I don't have the Shield Master feat, then I don't have that bonus action to take.

You start your turn with your move and your action. If you say "I want to shove that Orc and then hit it with my sword" at level 4 with the Shield Master feat, then the DM should say "well you can't actually do that because the shove consumes your action".

That's what the DM should say in your game. In my game, the DM should say, "I assume you're trying to knock it prone? Roll a Strength (Athletics) check. <Rolls> Does that beat a 13?"

Specifically, until you've taken the Attack action, you don't have a bonus action to shove someone. That's the way bonus actions with triggers work, the trigger must be true or you simply don't have the bonus action yet.

Once the condition "you take the Attack action on your turn" becomes true, it is true of your whole turn, including the part of your turn that happened before you took the Attack action. You deal with this by verifying if it's true only at the time the shove is declared and resolved. I deal with it by verifying if it's true only when I know whether or not the Attack action will be taken. I can understand the reasons for doing it your way, but for a number of reasons I think my way is better.

At some point the player's description of what they'd like to do needs to be converted into actual game mechanics.

Mechanics are needed only if it matters for resolution. All that's needed to resolve an attempt to shove a creature is a contested Strength check.

 

[Asgorath]

Once the condition "you take the Attack action on your turn" becomes true, it is true of your whole turn, including the part of your turn that happened before you took the Attack action. You deal with this by verifying if it's true only at the time the shove is declared and resolved. I deal with it by verifying if it's true only when I know whether or not the Attack action will be taken. I can understand the reasons for doing it your way, but for a number of reasons I think my way is better.

That's not now conditions work. The trigger condition is true from the point at which it becomes true until the end of your turn. You can't go back in time and say the trigger was true before the triggering event happened. D&D 5E is a sequential turn-based game, and as a result, timing and order of events matters.

Again, let's apply your logic to other triggered events.

Natural Explorer: "If you are traveling alone, you can move stealthily at a normal pace." 3 in-game years from now, my character will travel alone for day. Therefore, this trigger has been satisfied, and retroactively applies until the beginning of time because the rule doesn't explicitly say my turn. Thus, this rule says my character can simply move stealthily at full pace, period.

Shield Master: "If you are subjected to an effect that allows you to make a Dexterity saving throw to take only half damage, you can use your reaction to take no damage if you succeed on the saving throw, interposing your shield between yourself and the source of the effect." I can use my shield to take zero damage any time I succeed on a saving throw, because at some point in the future I'll make a Dex save. Because I'll make that Dex save at some point in the future, the triggering condition is true and thus I can retroactively get the benefit of this feature.

As you can see, this is nonsense, right?

 

[Arial Black]

So I just wanted to say, this thread was worth it. It's at least changed my opinion. I used to be anti-instantaneous actions and now I'm for them. Surprisingly that also means shield master can get triggered before the attacks from the attack action are taken which is completely different than my original position. So for those that offered interesting insights instead of just back and forth arguments, thank you!

You're welcome!

I've been catching up on reading the posts I've missed for a couple of days, and see a whole sequence of you and Epithet agreeing, while arguing with each other as if you disagree! Glad it's sorted now.

Incidentally, it speaks very well of you that you were convinced by logic, especially when it sometimes seems that the Internet is full of those whose pride prevents them from admitting they were ever wrong. Well done!

 

[epithet]

...
Natural Explorer: "If you are traveling alone, you can move stealthily at a normal pace." 3 in-game years from now, my character will travel alone for day. Therefore, this trigger has been satisfied, and retroactively applies until the beginning of time because the rule doesn't explicitly say my turn. Thus, this rule says my character can simply move stealthily at full pace, period.

Shield Master: "If you are subjected to an effect that allows you to make a Dexterity saving throw to take only half damage, you can use your reaction to take no damage if you succeed on the saving throw, interposing your shield between yourself and the source of the effect." I can use my shield to take zero damage any time I succeed on a saving throw, because at some point in the future I'll make a Dex save. Because I'll make that Dex save at some point in the future, the triggering condition is true and thus I can retroactively get the benefit of this feature.

As you can see, this is nonsense, right?

It is absolutely nonsense. You've tried the same nonsensical Natural Explorer example about a half dozen times now, and it just makes everything else you say in a post seem nonsensical by proximity. You should choose a better example, because this one doesn't help your argument at all.

Read the Natural Explorer feature again. Note the verb tense. It says, "If you are travelling..." there, doesn't it? It doesn't say "If you will travel" or "If you have travelled" or "If you have bought a ticket and plan to depart on Tuesday." The benefit of that feat must be concurrent with the trigger. Must happen at the same time as the trigger. I don't know how else to say it... you must, at the time that you "move stealthily at a normal pace," be actively travelling alone. What about this is difficult for you? It's in plain language, and is (as far as I can tell) unambiguous.

Regarding the saving throw benefit of Shield Master, being "subjected to an effect that allows you to make a Dexterity saving throw" is an objective event that takes place in the fictional world of the game, and as such it occupies a distinct place in the sequence of events on a turn. It is therefore unlike the Attack Action taking place "on your turn," which, as it says on the tin, occupies a unit of time consisting of "your turn." So, while the Shield Master shove must take place at a time concurrent with the Attack Action (ie, on your turn,) the saving throw benefit of the same feat must take place at a time concurrent with (or instantly and immediately following) being "subjected to an effect that allows you to make a saving throw," which is an instant or moment within your turn. The reason the saving throw benefit must be concurrent with or immediate to the trigger is that it is a Reaction, which "is an instant response to a trigger" as opposed to a Bonus Action, which you "choose when to take ... during your turn."

 

[Arial Black]

Disengage can absolutely be read as the DM giving you a token that says "your movement no longer provokes OAs" which they take back at the end of your turn. Dodge can absolutely be read as the DM giving you a token that says "attacks against you have disadvantage" that they take back at the start of your next turn.

And the Attack action can equally be read as the DM giving you a token that says "you may execute all your allowed weapon attacks between now and the end of your turn"!

The fallacy you're making is Special Pleading. Actions either are ALL 'instantaneous, with ongoing effects', OR they ALL 'have a duration'. Saying that some work one way and some work another, without written rules, is Special Pleading!

These all have very clear timing in the text, specifically talking about when the effect ends. The Attack action has no such language, so I'm not following how you think this lets you just take your attacks at some point in the future.

NONE of them have language that tell you whether the Action itself lasts for its duration or is instantaneous with ongoing effects!

The rules are silent on this issue. This is indicative that 'when an Action ends', as opposed to 'when the effects of an Action ends', was not considered to matter as far as the rules are concerned. This leads to the conclusion that 'Actions are indivisible' was NEVER part of the design when written, and has only become a thing post hoc in order for JC to justify his change of heart.

Because, if it did matter, they would have made it an actual rule!